GitHub Repository

Best static AI text humanizer. Two research-grounded skills that work in any LLM (Claude, ChatGPT, Gemini, Codex): humanize beats perplexity-based detectors, ai-check produces forensic scoring with evidence-quoted flags. Nine levers, 50+ peer-reviewed sources, 2024-2026 detection literature.

78 starsHTML

Humanize – Best static AI text humanizer, based on 50 research papers

Name: Humanize – Best static AI text humanizer, based on 50 research papers
Availability: InStock
Author: gokhalh

by gokhalh·Jul 7, 2026·2 points·0 comments

Visit Project View on HN

AI Analysis

●MidNiche Gem

Research-backed prompt skills for AI humanization when GPTZero and others already exist.

Strengths

•Nine specific humanization levers grounded in 50+ peer-reviewed detection papers.
•Works across any LLM agent without model-specific integration.
•Includes ai-check skill for forensic scoring with evidence-quoted flags.

Weaknesses

•AI text humanization is a crowded category with many established tools.
•Rule-based approach has acknowledged ceiling against learned classifiers.

Post Description

I thought humanizing AI writing was straightforward. Just delete the em dashes, adjust the tone, and remove a few obvious words.

However, as I "delved" into research on AI versus human writing, I discovered there was way more to it! Factors like sentence rhythm, repetition, hedge words, paragraph structure, punctuation habits, synonym cycling, and the "helpful assistant" voice reveal distinct patterns.

If you use AI for writing, I’d love your take!

The project is open source. Feedback, issues, ideas, and contributions are very welcome.